content-recommendation

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from ServiceNow incidents and cases (short descriptions, descriptions, and comments). This creates a surface for indirect prompt injection if an attacker-controlled ticket contains malicious instructions aimed at the agent.
  • Ingestion points: Data is retrieved via SN-Query-Table and REST API calls to the incident and sn_customerservice_case tables in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating ticket content into the recommendation logic.
  • Capability inventory: The skill possesses the capability to modify ServiceNow records (SN-Update-Record), add work notes (SN-Add-Work-Notes), and execute shell commands (Bash).
  • Sanitization: No specific sanitization of input data from ServiceNow fields is defined in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 01:34 PM
Security Audit — agent-trust-hub — content-recommendation